How do I install the Herdr skill?

Run npx skillmds@latest add kentoje/herdr-2 in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Herdr safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Herdr?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Herdr free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

kentoje (@kentoje) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Herdr by kentoje — Agent Skill

Name: Herdr
Author: kentoje

File contents

herdr

Before using this skill, check that HERDR_ENV=1. If it is not 1, say that the session is not inside a herdr-managed pane and stop. Do not inspect or control a focused herdr pane from outside herdr.

herdr exposes workspaces, tabs, and panes. Each pane is a real terminal running a shell, agent, server, or log stream. Agent status is idle, working, blocked, done, or unknown; done means the finished pane has not necessarily been inspected.

IDs are live-session identifiers: workspace IDs look like 1, tab IDs like 1:2, and pane IDs like 1-3. IDs can compact after closures. Always reread list or create/split responses instead of guessing an old ID.

Discover and inspect

herdr pane list
herdr workspace list
herdr tab list --workspace 1
herdr pane read 1-1 --source recent --lines 50

pane read supports visible, recent, and recent-unwrapped; the unwrapped form matches what wait output searches.

Tabs and workspaces

herdr tab create --workspace 1 --label "logs"
herdr tab rename 1:2 "logs"
herdr tab focus 1:2
herdr tab close 1:2
herdr workspace create --cwd /path/to/project --label "api"
herdr workspace focus 2
herdr workspace rename 1 "api"
herdr workspace close 2

Use --no-focus on create operations when the current context must remain focused.

Split, run, and coordinate

Split a pane, parse the returned result.pane.pane_id, then run a command:

herdr pane split 1-2 --direction right --no-focus
herdr pane run 1-3 "npm run dev"
herdr pane send-text 1-3 "echo ready"
herdr pane send-keys 1-3 Enter
herdr pane close 1-3

Use live IDs from the split response; do not assume the example ID remains valid.

Wait for output or an agent status rather than polling:

herdr wait output 1-3 --match "server.*ready" --regex --timeout 30000
herdr wait agent-status 1-1 --status done --timeout 60000
herdr pane read 1-3 --source recent-unwrapped --lines 40

An output wait timeout exits with status 1; inspect the pane before deciding whether to retry.

Safe recipes

For a server: split with --no-focus, run it in the new pane, wait for a specific readiness line, then read recent output. For another agent: list panes, wait for its target status, and read its recent output. Close temporary panes, tabs, and workspaces when finished.

---name: herdr-2description: Control the herdr terminal multiplexer through its CLI, including workspaces, tabs, panes, agents, and output waits. Use only when `HERDR_ENV=1` indicates the current session is inside herdr.---# herdrBefore using this skill, check that `HERDR_ENV=1`. If it is not `1`, say that the session is not inside a herdr-managed pane and stop. Do not inspect or control a focused herdr pane from outside herdr.herdr exposes workspaces, tabs, and panes. Each pane is a real terminal running a shell, agent, server, or log stream. Agent status is `idle`, `working`, `blocked`, `done`, or `unknown`; `done` means the finished pane has not necessarily been inspected.IDs are live-session identifiers: workspace IDs look like `1`, tab IDs like `1:2`, and pane IDs like `1-3`. IDs can compact after closures. Always reread list or create/split responses instead of guessing an old ID.## Discover and inspect```bashherdr pane listherdr workspace listherdr tab list --workspace 1herdr pane read 1-1 --source recent --lines 50````pane read` supports `visible`, `recent`, and `recent-unwrapped`; the unwrapped form matches what `wait output` searches.## Tabs and workspaces```bashherdr tab create --workspace 1 --label "logs"herdr tab rename 1:2 "logs"herdr tab focus 1:2herdr tab close 1:2herdr workspace create --cwd /path/to/project --label "api"herdr workspace focus 2herdr workspace rename 1 "api"herdr workspace close 2```Use `--no-focus` on create operations when the current context must remain focused.## Split, run, and coordinateSplit a pane, parse the returned `result.pane.pane_id`, then run a command:```bashherdr pane split 1-2 --direction right --no-focusherdr pane run 1-3 "npm run dev"herdr pane send-text 1-3 "echo ready"herdr pane send-keys 1-3 Enterherdr pane close 1-3```Use live IDs from the split response; do not assume the example ID remains valid.Wait for output or an agent status rather than polling:```bashherdr wait output 1-3 --match "server.*ready" --regex --timeout 30000herdr wait agent-status 1-1 --status done --timeout 60000herdr pane read 1-3 --source recent-unwrapped --lines 40```An output wait timeout exits with status `1`; inspect the pane before deciding whether to retry.## Safe recipesFor a server: split with `--no-focus`, run it in the new pane, wait for a specific readiness line, then read recent output. For another agent: list panes, wait for its target status, and read its recent output. Close temporary panes, tabs, and workspaces when finished.

Frequently asked questions

Herdr